marketing-copywriting

Pass

Audited by Gen Agent Trust Hub on Sep 22, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection due to its handling of untrusted data sources.
  • Ingestion points: The skill ingests untrusted user requests (e.g., instructions to "write copy for," "rewrite this page") and explicitly directs the agent to read local context files such as .agents/product-marketing.md, .claude/product-marketing.md, and product-marketing-context.md.
  • Boundary markers: The instructions lack explicit delimiters or "ignore embedded instructions" warnings for the content read from these external project files or user-provided prompts.
  • Capability inventory: While the primary purpose is text generation, the skill relies on file-reading tools. As the platform allows tool execution by default, these ingestion points could be manipulated if the context files contain malicious instructions.
  • Sanitization: No sanitization, escaping, or validation of external content is specified before interpolation into the agent's working context.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 22, 2026, 02:56 PM
Security Audit — agent-trust-hub — marketing-copywriting